You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hudi.apache.org by "Sagar Sumit (Jira)" <ji...@apache.org> on 2023/08/28 11:19:00 UTC
[jira] [Closed] (HUDI-4631) Enhance retries for failed writes w/ write conflicts in a multi writer scenarios
[ https://issues.apache.org/jira/browse/HUDI-4631?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Sagar Sumit closed HUDI-4631.
-----------------------------
Resolution: Fixed
> Enhance retries for failed writes w/ write conflicts in a multi writer scenarios
> --------------------------------------------------------------------------------
>
> Key: HUDI-4631
> URL: https://issues.apache.org/jira/browse/HUDI-4631
> Project: Apache Hudi
> Issue Type: Improvement
> Components: multi-writer
> Reporter: sivabalan narayanan
> Assignee: Sagar Sumit
> Priority: Blocker
> Labels: pull-request-available
> Fix For: 1.0.0
>
>
> lets say there are two writers from t0 to t5. so hudi fails w2 and succeeds w1. and user restarts w2 and for next 5 mins, lets say there are no other overlapping writers. So the same write from w2 will now succeed. so, whenever there is a write conflict and pipeline fails, all user needs to do is, just restart the pipeline or retry to ingest the same batch.
>
> Ask: can we add retries within hudi during such failures. Anyways, in most cases, users just restart the pipeline in such cases.
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)